Mlops (2026)

If you want to learn MLOps (Machine Learning Operations), I’d follow a mix of practitioners who build production ML systems, educators who explain the fundamentals, and engineers who share real-world lessons.

The key is to learn more than model training: MLOps is about deploying models, automating pipelines, monitoring performance, managing data and model versions, and running ML systems reliably in production.

Who to follow for MLOps

1. Chip Huyen

Best overall

Learn how to design and operate reliable ML systems in production. Excellent for understanding system architecture, deployment trade-offs, data pipelines, and monitoring.

Follow: Website & blog · LinkedIn

2. Goku Mohandas

Best for hands-on learning.

His Made With ML material teaches practical ML engineering, testing, reproducibility, deployment, and production workflows.

Follow: Made With ML · GitHub

3. Eugene Yan

Best for real-world engineering patterns.

Read his explanations of ML system design, recommendation systems, evaluation, and how teams put models into production.

Follow: Website & writing · LinkedIn

4. Demetrios Brinkmann

Best for learning from practitioners.

His MLOps community and MLOps Coffee Sessions podcast feature engineers discussing real production challenges, infrastructure choices, and lessons learned.

Follow: MLOps Community · Podcast

5. Hamel Husain

Best for practical ML and LLM evaluation.

Particularly useful if you're interested in evaluating AI applications, testing model behavior, building reliable LLM systems, and developing engineering workflows.

Follow: GitHub · LinkedIn

These recommendations cover complementary perspectives on ML systems, hands-on engineering, and production practice. Why MLOps Engineers Need to Think Like ML Researchers (And Here's Where to Start) 🧠22 Notable MLOps Experts & Engineers (2026)MLOps Community | Learn, Meet & Grow in Real-World MLOpsBuild and Deploy MLOps

Three communities and resources worth following

MLOps Community

Talks, newsletters, podcasts, and discussions with people running ML systems in production.

Join the community

DataTalks.Club - MLOps Zoomcamp

A practical learning path for experiment tracking, orchestration, deployment, monitoring, and infrastructure.

Explore courses and projects

Full Stack Deep Learning

Learn the broader lifecycle of production ML, including data management, deployment, testing, and maintaining AI products.

Free course materials

For hands-on tooling, also keep an eye on MLflow for experiment tracking and model lifecycle management, and Kubeflow for ML workflows and pipelines. CommunityCommunity & MeetingsGitHub - visenger/awesome-mlops: A curated list of references for MLOps

My recommended order

1. Start with Chip Huyen to develop the right mental models.
2. Work through Made With ML or MLOps Zoomcamp to build practical skills.
3. Follow Eugene Yan to understand engineering decisions in real systems.
4. Listen to MLOps Coffee Sessions to hear how experienced engineers solve production problems.
5. Build one end-to-end project: train a model, track experiments, deploy an API, automate testing, and monitor it.

One question so I can tailor the recommendations: What is your current background?

To truly master MLOps (Machine Learning Operations) and LLMOps, it is best to follow a mix of production-focused practitioners, prolific authors, community builders, and hands-on newsletters.

1. Key Thought Leaders & Authors to Follow

• Chip Huyen (Co-founder of Claypot AI, author of Designing Machine Learning Systems): Widely considered one of the definitive voices in MLOps. Her writing and talks break down the infrastructure, data challenges, and architectural patterns required to take ML to production. A list of AI leaders and communities you can follow - GitHub

• Demetrios Brinkmann (Founder of the MLOps Community): Through his extensive network, he talks to hundreds of platform engineers and ML practitioners daily. Following him gives you a pulse on what companies are actually running in production.

• Lilian Weng (Head of Safety / Applied Research at OpenAI): Though her focus leans heavily toward safety and applied research, her technical blog posts are legendary for explaining complex system mechanics, evaluation, and scaling. Top 40 Machine Learning Influencers on Twitter in 2026

• Hamel Husain (AI/ML Engineer & Consultant): Excellent resource for practical LLMOps, fine-tuning, evals, and developer workflows. His insights bridge the gap between software engineering and machine learning.

2. Essential Newsletters & Communities

• The MLOps Community Newsletter & Slack: The premier hub for MLOps engineers. Joining their Slack and reading their weekly digest exposes you to real-world deployment troubleshooting and open-source tooling discussions.

• Latent Space (by swyx and Alessio Fanelli): Essential reading/listening for AI engineers. It focuses intensely on LLM systems, evals, inference optimization, and production deployment tradeoffs. The best LLM and MLOps newsletters for engineers - daily.dev

• TheSequence : Great for turning fast-moving AI research and infrastructure updates into digestible engineering context.

3. Top Podcasts to Add to Your Queue

• Agentic Conversations (formerly The MLOps Podcast / MLOps Community): Hosted by Demetrios Brinkmann, this is the go-to long-form interview show featuring practitioners solving gritty production, orchestration, and infrastructure problems. 7 Best MLOps Podcasts to Listen to in 20267 Best MLOps Podcasts to Listen to in 2026

• Practical AI (Hosted by Chris Benson and Daniel Whitenack): Focuses heavily on applied AI, edge AI, and MLOps systems that actually reach production.

Where should you start? If you want a foundational text, pick up Chip Huyen’s book. If you want day-to-day engagement, join the MLOps Community Slack and subscribe to Latent Space. The best LLM and MLOps newsletters for engineers - daily.dev

Are you looking to focus more on traditional ML pipelines (monitoring, data drift, feature stores) or the newer LLMOps/Agentic infrastructure side?

To master MLOps (Machine Learning Operations), you should follow practitioners, researchers, and creators who actively build, write about, and architect production-grade machine learning systems.

The top experts, writers, and communities to follow include:

1. Key Authors & Practitioners

---

** Chip Huyen :** Co-founder of Claypot AI and author of Designing Machine Learning Systems. Her MLOps Guide by Chip Huyen is widely considered a gold-standard curriculum mapping out everything from feature stores to monitoring. Follow her insights on her blog and X/Twitter.

---

Demetrios Brinkmann : The force behind MLOps.community. He hosts regular podcasts, interviews with production ML engineers, and community roundtables. Following the MLOps.community gives you direct access to real-world war stories from practitioners at top tech companies.

---

Nathan Lambert : Known for his deep technical write-ups on RLHF, fine-tuning, and LLM operational pipelines via his Interconnects Newsletter. Excellent for keeping up with the operational side of modern generative AI.

---

Andrej Karpathy : While broader than just MLOps, his practical breakdowns of training infrastructure, autoregressive models, and engineering habits offer invaluable intuition for anyone scaling AI systems. Follow him on X/Twitter (@karpathy). MLOps guide - Chip Huyen15 AI Twitter Accounts to Follow for LLM Research and Insights

2. Essential Newsletters & Blogs

• • Marvelous MLOps (Substack): A brilliant curation of deep-dives, tool evaluations, and architectural patterns for production machine learning workflows.

• The DVC Blog (Data Version Control): Maintained by the team at Iterative, the DVC Blog provides hyper-practical, code-heavy tutorials on data pipelines, experiment tracking, and reproducibility.

3. Communities & Interactive Learning

• • MLOps.community Slack/Discord: Joining the MLOps.community platform connects you with thousands of engineers sharing troubleshooting tips, tool comparisons (like Kubeflow vs. MLflow vs. Metaflow), and career advice. MLOps guide - Chip Huyen

• Full Stack Deep Learning: A fantastic resource co-founded by industry veterans that teaches you how to build, deploy, monitor, and operate production ML systems end-to-end.

I’d start with Chip Huyen, Goku Mohandas, and Alexey Grigorev: they offer a useful mix of systems thinking, hands-on engineering, and structured MLOps practice.(huyenchip)(madewithml)(github.com)

People worth following

Person	What you’ll learn	Where to start
Chip Huyen	How to think about production ML systems: data pipelines, deployment requirements, serving, and the tradeoffs behind architectural decisions.	Her blog and MLOps guide, which lays out the learning path she would follow starting again. (huyenchip)(huyenchip)
Goku Mohandas	How to build an end-to-end ML application, including experiment tracking, testing, reproducibility, serving, CI/CD, and monitoring.	Made With ML - particularly useful if you prefer learning through code rather than social-media posts. (madewithml)(madewithml)
Alexey Grigorev	Practical production workflows, from experimentation to deploying and operating models.	DataTalks.Club’s MLOps Zoomcamp, where he is an instructor. The free course includes hands-on workshops and a final project. (github.com)
Elena Samuylova	What happens after deployment: model quality, data drift, monitoring, and evaluation.	Her articles on the Evidently AI blog, including monitoring blueprints and drift-detection experiments. (evidentlyai)
Demetrios Brinkmann	How practitioners approach real production problems across different teams and organizations.	The MLOps Community podcast and technical talks; Brinkmann founded the community. (mlops.community)(mlops.community)

How I’d use this list

Rather than following everyone equally, pick one source for each learning mode:

• For conceptual understanding: read Chip Huyen’s systems-design material.(huyenchip)(huyenchip)
• For implementation: work through either Made With ML or MLOps Zoomcamp and build the accompanying project.(github.com)(madewithml)
• For operational depth: read Elena Samuylova’s monitoring articles and listen to MLOps Community practitioner interviews.(evidentlyai)(mlops.community)

My suggested starting combination is Chip Huyen for the “why” and Goku Mohandas for the “how.” Their materials complement each other: one emphasizes system design, while the other connects the components in working production workflows.(huyenchip)(madewithml)

If you want to learn MLOps, follow a mix of open‑source pioneers, educators, and community builders. Some of the most influential voices include Matei Zaharia (Apache Spark, MLflow), Jeremy Lewi (Kubeflow), Chip Huyen (ML systems design), Abhishek Thakur (Kaggle Grandmaster turned educator), and Demetrios Brinkmann (MLOps Community).22 Notable MLOps Experts & Engineers (2026)

📌 Key People to Follow in MLOps

Expert	Known For	Why Follow
Matei Zaharia	Creator of Apache Spark & MLflow	Deep insights into ML lifecycle tools and scalable infrastructure.
Jeremy Lewi	Co‑founder of Kubeflow	Practical expertise in deploying ML workflows on Kubernetes.
Abhishek Thakur	4x Kaggle Grandmaster, author	Bridges competition ML with real‑world engineering and education.
Chip Huyen	ML systems design, author	Clear writing on production ML, system design, and scaling.
Ville Tuulos	Co‑founder of Outerbounds (Metaflow)	Focus on ML infrastructure and developer productivity.
Hamel Husain	MLOps automation	Shares practical automation strategies for ML pipelines.
Demetrios Brinkmann	Founder of MLOps Community	Hosts podcasts, events, and discussions with practitioners.
Alejandro Saucedo	Open‑source MLOps advocate	Strong voice on ethics, reproducibility, and open tooling.
Mike Del Balso	Co‑founder of Tecton (feature stores)	Insights into feature engineering and operational ML.
Aparna Dhinakaran	Co‑founder of Arize AI	Specializes in ML observability and monitoring.22 Notable MLOps Experts & Engineers (2026)

📚 Blogs & Communities

• MLOps Community Blog – Central hub for articles, podcasts, and events.

• Neptune AI Blog – Practical guides on experiment tracking and reproducibility.

• Evidently AI Blog – Focused on monitoring and evaluation of ML models.

• ZenML Blog – Tutorials on pipeline orchestration.

• ClearML Blog – Covers automation and scaling ML workflows.25 Best MLOps Blogs to Follow in 2026

🔑 How to Use These Resources

• LinkedIn & Twitter/X: Follow these experts for ongoing discussions, event announcements, and thought leadership.

• Podcasts & Newsletters: Subscribe to MLOps Community podcasts or newsletters like SwirlAI for curated updates.

• Hands‑on Learning: Many of these experts share open‑source projects (e.g., MLflow, Kubeflow, Metaflow). Contributing or experimenting with them is one of the fastest ways to learn.

⚠️ Things to Watch Out For

• Hype vs. Practice: MLOps is full of buzzwords. Prioritize voices who share practical case studies and open‑source code over vague futurism.

• Tool Overload: The ecosystem is fragmented. Focus on a few core tools (MLflow, Kubeflow, Metaflow, feature stores) before branching out.

• Community Value: Engage with communities like MLOps World or MLOps Community to learn from practitioners, not just influencers.Speakers - MLOps World

👉 Would you like me to curate a starter learning path (courses, blogs, and GitHub repos) so you can move from beginner to advanced in MLOps step by step?